[daip] Re: A question on flux density scales

[Eric Greisen]

papadop at phys.ethz.ch writes:
 > Hi again Eric,
 > 
 > I discovered a rather strange thing on the flux measurement of the 2ndary,
 > 
 > I do my editing and produce an SN table for freqid 2, and have also set the
 > fluxes of 1331+305 using SETJY with 'CALC' option  and that of 0713+438
 > with the option of putting it by hand (I have set I, Q, U, V = 0.55, 0, 
 > 0, 0).
 > 
 > Then I do a GETJY on the 2ndary 06465+44513 (after I have used SETJY to
 > set it to zero).
 > 
 > In the attachement you can see the results of doing this using a) both 
 > primary calibrators
 > b)  only 1331+305, and c) only 0713+438.
 > 
 > The fluxes I get for the 2ndary are disturbingly different.....,
 > 
 > I am at a loss....
 > 
 > 
 > again thanks for your help
 > 
 > Padelis
 > 
 > PS: The CALIB was run for all 3 cal sources with no UV restrictions (D 
 > array, K Band)
 >      the results you can see also in the attached file (I have chosen 
 > soltype 'R')
 > localh> CALIB1: Antenna 10 had     1 excess closure errors
 > localh> CALIB1: Antenna 11 had     1 excess closure errors
 > localh> CALIB1: Found         4792 good solutions
 > localh> CALIB1: Failed on        4 solutions
 > localh> CALIB1: Fraction of times having data > 2.5 rms from solution
 > localh> CALIB1: 0.51208 of the times had  0 -  2 percent outside 2.5 times rms
 > localh> CALIB1: 0.14976 of the times had  2 -  4 percent outside 2.5 times rms
 > localh> CALIB1: 0.03865 of the times had  4 -  6 percent outside 2.5 times rms
 > localh> CALIB1: 0.03382 of the times had  6 -  8 percent outside 2.5 times rms
 > localh> CALIB1: 0.03865 of the times had  8 - 10 percent outside 2.5 times rms
 > localh> CALIB1: 0.04348 of the times had 10 - 12 percent outside 2.5 times rms
 > localh> CALIB1: 0.01932 of the times had 12 - 14 percent outside 2.5 times rms
 > localh> CALIB1: 0.01932 of the times had 14 - 16 percent outside 2.5 times rms
 > localh> CALIB1: Appears to have ended successfully
 > ------------------------------------------------------------------------------
 > ------------------------------------------------------------------------------
 > 
 > WITH BOTH 1331+305, 0713+438
 > 
 > localh> CALIB1: localhost    31DEC04 TST: Cpu=       0.7  Real=       1
 > localh> GETJY1: Task GETJY  (release of 31DEC04) begins
 > localh> GETJY1:    Source:Qual      CALCODE IF  Flux (Jy)
 > localh> GETJY1: 06465+44513     :  0   A     1    3.98157 +/-   0.13750
 > localh> GETJY1:                              2    3.94069 +/-   0.13479
 > ----------------------------------------------------------------------------------
 > WITH ONLY 1331+305
 > 
 > localh> SETJY1: Appears to have ended successfully
 > localh> SETJY1: localhost    31DEC04 TST: Cpu=       0.0  Real=       0
 > localh> GETJY1: Task GETJY  (release of 31DEC04) begins
 > localh> GETJY1:    Source:Qual      CALCODE IF  Flux (Jy)
 > localh> GETJY1: 06465+44513     :  0   A     1    1.10579 +/-   0.06850
 > localh> GETJY1:                              2    1.10597 +/-   0.06705
 > localh> GETJY1: Appears to have ended successfully
 > -------------------------------------------------------------------------------
 > 
 > with ONLY 0713+438
 > 
 > localh> GETJY1: Task GETJY  (release of 31DEC04) begins
 > localh> GETJY1:    Source:Qual      CALCODE IF  Flux (Jy)
 > localh> GETJY1: 06465+44513     :  0   A     1    0.80867 +/-   0.01006
 > localh> GETJY1:                              2    0.80837 +/-   0.00849
 >
----------------------------------------------------------------------------

When you run GETJY, the fluxes are changed in the SU table and the
gains are changed in the SN table.  If you then run CALIB again, the
GETJY fluxes (rather than 1.0) are used.  If you run GETJY again using
the same SN table, then a different result will be found for the flux
and things changed further.  I suspect that you should back up and be
careful about the sequence in which you do things.

What flux do you expect for 06465+44513 ?  Note - if the 4 Jy is about
right and you just re-run GETJY then getting nearly 1.0 next time is
what one would expect.

You might want to use a source model for CALIB on 1331+305.  They are
available in aips (see CALDIR, CALRD).

Eric Greisen